The transmission of complex phenotypes such as type 2 diabetes is likely to reflect the actions and interactions of multiple genetic and environmental factors. Linkage studies designed to localize genes for type 2 diabetes have yield few regions with highly significant evidence for linkage and fewer still that have been replicated in additional studies. While these results are disappointing, they can hardly be considered surprising. The failure of the current generation of studies to demonstrate conclusive and reproducible localization of the genes contributing to type 2 diabetes likely results from the combined effects of inadequate sample sizes, incorrect or incomplete models for genetic analysis, and the genuine complexity of the genetic component to disease susceptibility. While it is widely recognized that increasing the size of the sample included in linkage studies will provide the power to detect and localize susceptibility loci with modesty or moderate effects, collection of family data is the most expensive and time-consuming aspect of linkage studies. In an effort to maximize the utility of data that already have been collected to map genes for type 2 diabetes, we have formed The International Type 2 Diabetes Linkage Analysis Consortium to combine existing data sets for linkage analysis. The specific aims are: (1) to conduct linkage analyses over all human autosomes and the X chromosome within and across the racial/ethnic groups represented in the combined Consortium data to identify regions most likely to contain type 2 diabetes susceptibility loci; and (2) to conduct additional genotyping in Consortium data in regions of interest identified in analyses of Consortium data and for mutations in diabetes susceptibility loci identified by us or others during the course of this study.